Content
47%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a well-organized catalogue stub: concise and cleanly sectioned, but it delegates all real instruction to an external repository without any executable steps. An agent reading it learns where the skill comes from and how to ask for it, but not how to do anything with it — the placeholder install command and absent usage guidance are the main weaknesses.
Suggestions
Replace the placeholder `open https://github.com/MiniMax-AI/skills` with a real, executable install command (e.g., git clone with the exact subdirectory and copy target for the skill bundle).
Add a minimal quick-start section showing one concrete PptxGenJS usage (a small Node.js snippet creating a basic deck) so the skill delivers value even before the upstream bundle is installed.
Include a verification step after installation (e.g., checking the skill directory lists the expected scripts/references) to close the workflow gap between 'install the bundle' and 'invoke the skill'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean with no padding or explanation of concepts Claude already knows. Minor trimming is possible: "What it does" repeats the frontmatter description verbatim, and the "> Curated from the MiniMax AI team" tagline adds little. This matches anchor 4 (efficient, minor instances that could be trimmed) rather than 5, where every token would earn its place. | 4 / 5 |
Actionability | There is no executable guidance for the skill's actual purpose — no PptxGenJS example, no install command, and the only code block is a placeholder ("# Inspect the upstream README for exact paths" with `open <url>`). It provides high-level hints with a specific URL and invocation instructions, which matches anchor 2; it is not 1 because it does include concrete pointers (upstream URL, skill name, trigger-based invocation). | 2 / 5 |
Workflow Clarity | A rough sequence is present (inspect upstream → install the bundle → ask the agent to invoke the skill), but the steps are poorly defined: no install command, no exact paths, and no verification that the skill is installed. Anchor 2 fits; it is not 3 because the steps are too underspecified to actually execute, which goes beyond merely missing checkpoints. | 2 / 5 |
Progressive Disclosure | The body is short, cleanly sectioned (What it does / Source / How to use), and its single external pointer (the upstream GitHub URL) is clearly signaled and one level deep with no buried or nested references. This matches anchor 4 (good structure, minor organization gaps); it is not 5 because the delegated content has no local anchor — a reader must leave the file to find anything. | 4 / 5 |
Total | 12 / 20 Passed |